ErrLookup › open-mmlab/mmdetection

open-mmlab/mmdetection

OpenMMLab Detection Toolbox and Benchmark · Python · 1,681 source files

Analyzed at cfd5d3a985 on 2026-08-27. 225 documented errors.

Code / MessageTypeSeverityTags
LoadImageFromFile is not found in the test pipeline
validation error mmdetection, inferencer, pipeline, config, loadimagefromfile
Visualization needs the "visualizer" termdefined in the conf
validation error mmdetection, inferencer, visualizer, config
Unsupported input type: {type(single_input)}
validation error mmdetection, inferencer, visualization, input-type, numpy
panopticapi is not installed, please install it by: pip inst
exception error mmdetection, panoptic-segmentation, missing-dependency, panopticapi
config must be a filename or Config object, but got {type(co
validation error mmdetection, config, typeerror, init-detector
Please run accumulate() first
exception error mmdetection, coco-eval, evaluation, state-machine, multiprocessing
The `file_client_args` is deprecated, please use `backend_ar
exception error mmdetection, dataset, deprecation, config-migration, backend-args
new classes {new_classes} is not a subset of classes {old_cl
validation error mmdetection, semantic-segmentation, dataset, metainfo, classes
palette does not match classes as metainfo is {self._metainf
validation error mmdetection, semantic-segmentation, palette, metainfo, visualization
The annotation file of Open Images Challenge should be a txt
validation error mmdet, dataset, annotation-file, file-extension, open-images
Invalid text mode "{self.text_mode}".
validation error mmdet, refcoco, text-mode, config, referring-expression
No sample in split "{self.split}".
validation error mmdet, refcoco, empty-dataset, split, annotations
sampler should be an instance of ``Sampler``, but got {sampl
validation error mmdet, sampler, batch-sampler, typeerror, pytorch
batch_size should be a positive integer value, but got batch
validation error mmdet, sampler, batch-size, valueerror, validation
dataset metainfo must contain `classes`
validation error mmdet, sampler, class-aware, metainfo, classes
only {num_videos} videos loaded,but {self.world_size} gpus w
validation error mmdet, tracking, distributed, world-size, video-sampler
The type of frame_range must be int or list.
validation error mmdet, video, frame-sampling, typeerror, config
img_border_value must be float or tuple with 3 elements.
validation error mmdet, transform, geometric, img-border-value, validation
Please run "pip install instaboostfast" to install instaboos
exception error mmdetection, instaboost, missing-dependency, augmentation, importerror
Please run "pip install instaboostfast" to install instaboos
exception error mmdetection, instaboost, lazy-import, runtime-importerror, augmentation
panopticapi is not installed, please install it by: pip inst
exception error mmdetection, panopticapi, panoptic-segmentation, missing-dependency, importerror
transformers is not installed, please install it by: pip ins
exception error mmdetection, transformers, huggingface, grounding-detection, missing-dependency
Invalid scale {scale}, must be positive.
validation error mmdetection, resize, valueerror, config-validation, scale
Scale must be a number or tuple of int, but got {type(scale)
validation error mmdetection, resize, typeerror, config-validation, type-mismatch
Invalid crop_type {crop_type}.
validation error mmdetection, random-crop, valueerror, config-validation, augmentation
imagecorruptions is not installed
exception error mmdet, dependency, imagecorruptions, robustness
albumentations is not installed
exception error mmdet, albumentations, dependency, augmentation
type must be a str or valid type, but got {type(obj_type)}
exception error mmdet, albumentations, config-validation, typeerror
Albu only supports horizontal boxes now
exception error mmdet, albumentations, bbox, data-structure
Albu only supports BitMap masks now
exception error mmdet, albumentations, instance-segmentation, polygon-masks
RandomCenterCropPad only supports bbox.
exception error mmdet, yolox, randomcentercroppad, segmentation
RandomCenterCropPad only support two testing pad mode:logica
exception error mmdet, yolox, pad-mode, config-validation
RandomAffine only supports bbox.
exception error mmdet, randomaffine, yolo, segmentation
results does not contain masks.
exception error mmdet, copypaste, masks, config
Not supported YouTubeVIS datasetversion: {dataset_version}
exception error mmdet, youtube-vis, dataset-version, config-validation
Unrecognized dataset: {dataset}
exception error dataset-name, mmdetection, config, validation
dataset must a str, but got {type(dataset)}
exception error type-error, mmdetection, dataset-name
Unrecognized mode, only "area" and "11points" are supported
exception error mean-ap, mmdetection, invalid-argument
tpfp_fn has to be a function or None, but got {tpfp_fn}
exception error mean-ap, mmdetection, callable-validation
panopticapi is not installed, please install it by: pip inst
exception error panoptic, missing-dependency, mmdetection, evaluation
In the image with ID {} segment with ID {} is presented in P
exception error panoptic, data-consistency, mmdetection, evaluation
In the image with ID {} segment with ID {} has unknown categ
exception error panoptic, category-id, data-consistency, mmdetection
In the image with ID {} the following segment IDs {} are pre
exception error panoptic, data-consistency, mmdetection, evaluation
unknown iouType for iou computation
exception error ytvis, video-instance-segmentation, iou-type, mmdetection
Please run accumulate() first
exception error ytvis, api-misuse, eval-pipeline, mmdetection
iouType not supported
exception error ytvis, iou-type, invalid-argument, mmdetection
device must be 'cpu' or 'gpu', but got {device}
exception error video-metric, collect-device, mmdetection, distributed
GPU collecting has not been supported yet
exception error video-metric, gpu-unsupported, mmdetection, distributed
Failed to import `cityscapesscripts`.Please try to install o
exception error cityscapes, missing-dependency, mmdetection, metric-init
The `file_client_args` is deprecated, please use `backend_ar
exception error deprecation, file-client, mmdetection, migration
COCOEvalCap is not installed, please install it by: pip inst
exception error captioning, missing-dependency, mmdetection, pycocoevalcap
metric should be one of 'bbox', 'segm', 'proposal', 'proposa
validation error mmdetection, coco, metric-validation, config-error
The `file_client_args` is deprecated, please use `backend_ar
validation error mmdetection, deprecation, file-client, migration
{metric} is not in results
validation error mmdetection, coco, metric-mismatch, missing-predictions
metric item "{metric_item}" is not supported
validation error mmdetection, coco, metric-validation, logging
panopticapi is not installed, please install it by: pip inst
validation error mmdetection, panoptic-segmentation, missing-dependency, pip
The `file_client_args` is deprecated, please use `backend_ar
validation error mmdetection, deprecation, panoptic-segmentation, migration
no prediction for the image with id: {img_id}
validation critical mmdetection, panoptic-segmentation, pq, data-mismatch
metric should be one of 'MR', 'AP', 'JI',but got {metric}.
validation error mmdetection, crowdhuman, metric-validation, config-error
The `file_client_args` is deprecated, please use `backend_ar
validation error mmdetection, crowdhuman, deprecation, migration
Please run "pip install jsonlines" to install this package.
validation error mmdetection, odvg, missing-dependency, pip, jsonl
The `file_client_args` is deprecated, please use `backend_ar
validation error mmdetection, deprecation, config, fileio
{k} is not a valid recall threshold
validation error mmdetection, flickr30k, recall, argument-validation
Package lvis is not installed. Please run "pip install git+h
validation error mmdetection, lvis, missing-dependency, evaluation
metric should be one of 'bbox', 'segm', 'proposal', 'proposa
validation error mmdetection, lvis, config, argument-validation
The `file_client_args` is deprecated, please use `backend_ar
validation error mmdetection, deprecation, lvis, config, fileio
trackeval is not installed,please install it by: pip install
validation error mmdetection, tracking, mot, trackeval, missing-dependency, numpy
metric must be a list or a str.
validation error mmdetection, tracking, mot, type-error, config
metric {metric} is not supported.
validation error mmdetection, tracking, mot, argument-validation, config
{metric} is not in results
validation error mmdet, evaluation, coco-metric, open-vocabulary, config
metric item "{metric_item}" is not supported
validation error mmdet, evaluation, coco-metric, metric-items, validation
metric must be a list or a str.
validation error mmdet, reid, typeerror, config, validation
metric {metric} is not supported.
validation error mmdet, reid, unsupported-metric, config
metrics {iou_metrics} is not supported. Only supports mIoU/m
validation error mmdet, semantic-segmentation, iou-metric, config, validation
metric should be one of 'recall', 'mAP', but got {metric}.
validation error mmdet, voc, evaluation, config
metric should be 'youtube_vis_ap', but got {metric}.
validation error mmdet, youtube-vis, video-instance-seg, config
metric item "{metric_item}" is not supported
validation error mmdet, youtube-vis, metric-items, evaluation
frozen_stages must be in range(-1, len(arch_setting) + 1). B
validation error mmdet, backbone, cspdarknet, frozen-stages, config
frozen_stages must be in range(-1, len(arch_setting) + 1). B
validation error mmdet, backbone, cspnext, frozen-stages, config
invalid depth {depth} for darknet
validation error mmdet, backbone, darknet, invalid-depth, config
pretrained must be a str or None
validation error mmdet, darknet, pretrained-weights, init-cfg, config
`init_cfg` must contain the key "type"
validation error mmdet, detectors-resnet, init-cfg, pretrained-weights, config
pretrained must be a str or None
validation error mmdet, detectors-resnet, pretrained-weights, init-weights, migration
the item in out_indices must in range(0, {len(self.layer_set
validation error mmdet, efficientnet, out-indices, backbone, config
frozen_stages must be in range(0, {len(self.layer_setting) +
validation error mmdet, efficientnet, frozen-stages, backbone, config
NUM_BRANCHES({num_branches}) != NUM_BLOCKS({len(num_blocks)}
validation error mmdet, hrnet, backbone, config-validation
NUM_BRANCHES({num_branches}) != NUM_CHANNELS({len(num_channe
validation error mmdet, hrnet, backbone, config-validation
NUM_BRANCHES({num_branches}) != NUM_INCHANNELS({len(in_chann
validation error mmdet, hrnet, backbone, config-validation
pretrained must be a str or None
validation error mmdet, hrnet, pretrained-weights, init-cfg, migration
pretrained must be a str or None
exception error mmdet, mobilenetv2, config, init-cfg, pretrained
out_indices must be a subset of range(0, 8). But received {o
exception error mmdet, mobilenetv2, out-indices, validation, config
frozen_stages must be in range(-1, 8). But received {frozen_
exception error mmdet, mobilenetv2, frozen-stages, validation, fine-tuning
pretrained must be a str or None
exception error mmdet, pvt, pretrained, init-cfg, config
Expect "arch" to be either a string or a dict, got {type(arc
exception error mmdet, regnet, arch-settings, config, validation
pretrained must be a str or None
exception error mmdet, regnet, pretrained, init-cfg
invalid depth {depth} for resnet
exception error mmdet, resnet, depth, arch-settings, config
pretrained must be a str or None
exception error mmdet, resnet, pretrained, init-cfg, deprecation
pretrained must be a str or None
exception error mmdet, ssd-vgg, pretrained, init-cfg
pretrained must be a str or None
exception error mmdet, swin-transformer, pretrained, init-cfg
Output of `cast_data` should be a dict or a tuple with input
exception error mmdet, data-preprocessor, batch-structure, type-error
skimage is not installed, please install it
exception error mmdet, missing-dependency, scikit-image, installation
Please run "pip install openmim" and run "mim install mmpret
exception error mmdet, mmpretrain, missing-dependency, reid, tracking
num_classes={num_classes} is too small
exception error mmdet, anchor-head, num-classes, config, validation
DETR do not build sampler.
exception error mmdet, detr, train-cfg, sampler, config
Unknown anchor type: {self.anchor_type}.Please use `anchor_f
exception error
{kernel} kernel is not supported in matrix nms!
exception error mmdet, nms, config, not-implemented
Unknown query_scale_type: {}
exception error mmdet, dab-detr, transformer, config, not-implemented
If you want to reduce GPU memory usage,
exception error mmdet, deformable-detr, fairscale, dependency, gpu-memory
There is not post_norm in {name}
exception error mmdet, deformable-detr, transformer, config
If you want to reduce GPU memory usage,
exception error mmdet, detr, fairscale, dependency, gpu-memory
If you want to reduce GPU memory usage,
exception error mmdet, grounding-dino, fairscale, dependency, gpu-memory
There is not post_norm in {name}
exception error mmdet, grounding-dino, transformer, config
Unknown pos_tensor shape(-1):{}
exception error mmdet, transformer, positional-encoding, shape-mismatch
The size of the 2D attn_mask is not correct.
exception error mmdet, attention, attn-mask, shape-mismatch
The size of the 3D attn_mask is not correct.
exception error mmdet, attention, attn-mask, shape-mismatch
attn_mask's dimension {} is not supported
exception error mmdet, attention, attn-mask, shape-mismatch
avg_factor can not be used with reduction="sum"
validation error mmdet, loss, reduction, avg-factor
module must be a str or a list.
exception error mmdetection, mot, config, type-error
Invalid mode "{mode}". Only supports loss, predict and tenso
exception error mmdetection, mot, forward-mode
_forward function (namely 'tensor' mode) is not supported no
exception error mmdetection, mot, not-implemented
Please train `detector` and `reid` models firstly, then
exception error mmdetection, tracking, sort, deepsort
Please run "pip install openmim" and run "mim install mmpret
exception critical mmdetection, mmpretrain, reid, missing-dependency
neck inputs should be tuple or torch.tensor
exception error mmdetection, reid, neck, type-error
Please run "pip install openmim" and run "mim install mmpret
exception critical mmdetection, mmpretrain, reid, missing-dependency
Please choose at least one loss in triplet loss and cross en
validation error mmdetection, reid, loss, config
The num_classes must be a current number, if there is cross
validation error mmdetection, reid, config, type-error
The last dim of `cls_scores` should equal to `num_classes` o
validation error mmdetection, bbox-head, shape-mismatch, config
Invalid upsample method {}, accepted methods are "deconv", "
validation error mmdetection, mask-head, config
grid_points must be a square number
validation error mmdetection, grid-head, config
Only square RoIs are supporeted in Grid R-CNN
validation error mmdetection, grid-head, config
pretrained must be a str or None
validation error mmdetection, reslayer, shared-head, config
Please run "pip install scipy" to install scipy first.
error_code critical mmdetection, detr, hungarian-assigner, scipy, missing-dependency
center_offset should be in range [0, 1], {center_offset} is
validation error mmdetection, anchor-generator, config
Either scales or octave_base_scale with scales_per_octave sh
validation error mmdetection, anchor-generator, config
basesize_ratio_range[0] should be either 0.15or 0.2 when inp
validation error mmdetection, ssd, anchor-generator, config-validation
When not setting min_sizes and max_sizes,basesize_ratio_rang
validation error mmdetection, ssd, anchor-generator, config-validation
Only support 300 or 512 in SSDAnchorGenerator when not setti
validation error mmdetection, ssd, anchor-generator, input-size
sscikit-learn is not installed, please insta
error_code error mmdetection, tracking, missing-dependency, scikit-learn
sscikit-learn is not installed, please insta
error_code error mmdetection, tracking, missing-dependency, scipy, kalman-filter
Invalid flipping direction '{flip_direction}'
validation error mmdetection, test-time-augmentation, flip, mask
kwargs value must both equal
validation error mmdetection, tracking, shape-mismatch
lap is not installed, please install it by:
error_code error mmdetection, tracking, bytetrack, missing-dependency, lap
lap is not installed, please install it by:
error_code error mmdetection, tracking, ocsort, missing-dependency, lap
motmetrics is not installed, please install
error_code error mmdetection, tracking, sort, missing-dependency, motmetrics
motmetrics is not installed, please install
error_code error mmdetection, tracking, strongsort, missing-dependency, motmetrics
Regression head is not supported yet.
validation error mmdetection, tracking, qdtrack, not-implemented
Regression head is not supported yet.
validation error mmdetection, tracking, roi-head, not-implemented
Unsupported {type(mask)} data type
validation error mmdetection, mask, type-validation
Only supports dict or list or Tensor, but get {type(results)
validation error mmdetection, dense-head, topk, type-validation
The hidden size ({config.hidden_size}) is not a multiple of
validation error mmdetection, bert, attention, config-validation
boxes should be Tensor, ndarray, or Sequence, but got {type(
validation error mmdetection, bbox, type-validation
box_list should not be a empty list.
validation error mmdetection, bbox, empty-input
box type {name} has been registered
validation error mmdetection, registry, duplicate-registration
please run pip install seaborn
exception error python, dependency, visualization, mot, seaborn
motmetrics is not installed, please install it by: pip insta
exception error python, dependency, visualization, misleading-message, seaborn
Invalid type for palette: {type(palette)}
validation error python, validation, palette, typeerror, visualization
dataset_meta or class names are not saved in the checkpoint'
console warning python, warning, checkpoint, class-names, inferencer
Checkpoint is not loaded, and the inference result is calcul
console warning python, warning, checkpoint, random-init, inferencer
weights is None, use COCO classes by default.
console info python, warning, class-names, inferencer
palette does not exist, random is used by default. You can a
console info python, warning, palette, inferencer, visualization
Currently does not support saving datasample when return_dat
console warning python, warning, inferencer, serialization, datasamples
checkpoint is None, use COCO classes by default.
console warning python, warning, checkpoint, init-detector, random-init
dataset_meta or class names are not saved in the checkpoint'
console warning python, warning, checkpoint, class-names, init-detector
palette does not exist, random is used by default. You can a
console info python, warning, palette, init-detector, visualization
dataset_meta or class names are missed, use None by default.
console info python, warning, tracking, dataset-meta, mot
mmpycocotools is deprecated. Please install official pycocot
console warning python, warning, pycocotools, deprecated-dependency, coco
Cache files can not be saved automatically! To speed uploadi
console warning python, warning, crowdhuman, cache, filesystem, dataset
mmlvis is deprecated, please install official lvis-api by "p
console warning python, warning, lvis, deprecated-dependency, dataset
gt_masks is already contained in results, so paste_by_box is
console info python, warning, copy-paste, augmentation, segmentation
The show is True, it means that only the prediction results
console info python, warning, visualization, hook, show-mode
{metric.__class__.__name__} got empty `self.results`.Please
console warning python, warning, evaluation, multi-dataset, empty-results
{self.__class__.__name__} got empty `self.results`. Please e
console warning mmdet, video-metrics, empty-results, evaluation, tracking
{self.__class__.__name__} got empty `self.results`. Please e
console warning mmdet, coco-video, empty-results, evaluation, tracking
Panoptic segmentation map will not be compressed. The dumped
console warning mmdet, panoptic-segmentation, dump-results, disk-usage, coco
mmlvis is deprecated, please install official lvis-api by "p
console warning mmdet, lvis, deprecated-package, dependency, evaluation
Pascal VOC2007 uses `11points` as default evaluate mode, but
console warning mmdet, voc, evaluation-protocol, map, config
Pascal VOC2012 uses `area` as default evaluate mode, but you
console warning mmdet, voc, evaluation-protocol, map, config
DeprecationWarning: pretrained is deprecated, please use "in
console warning mmdet, darknet, deprecated-argument, init-cfg, backbone
DeprecationWarning: pretrained is deprecated, please use "in
console info mmdet, swin-transformer, deprecation, pretrained-weights
No pre-trained weights for {self.__class__.__name__}, traini
console warning mmdet, swin-transformer, pretrained-weights, initialization
DeprecationWarning: `num_anchors` is deprecated, for consist
console info mmdet, anchor-head, deprecation, rename
DeprecationWarning: anchor_generator is deprecated, please u
console info mmdet, anchor-head, deprecation, rename
nms_cfg in test_cfg will be deprecated. Please rename it as
console warning mmdet, cornernet, config-migration, nms
You are calling `simple_test_bboxes` in `dense_test_mixins`,
console info mmdet, deprecation, dense-head, testing
You are calling `aug_test_bboxes` in `dense_test_mixins`, bu
console info mmdet, deprecation, tta, dense-head
The `in_channels` of SOLOv2MaskFeatHead and SOLOv2Head shoul
console warning mmdet, solov2, config, instance-segmentation
There is overlap between pos and neg sample.
console warning mmdet, yolo, training, target-assignment
Inputting a text that is too long will result in poor predic
console warning mmdet, glip, text-prompt, token-limit
Inputting a text that is too long will result in poor predic
console warning mmdet, glip, chunked-inference, token-limit
The unexpected output indicates an issue with named entity r
console warning mmdet, glip, ner, grounding
Inputting a text that is too long will result in poor predic
console warning mmdet, grounding-dino, chunked-inference, token-limit
The unexpected output indicates an issue with named entity r
console warning mmdet, grounding-dino, ner, grounding
The `num_classes` should be 1 in RPN, but get {rpn_head_num_
console warning mmdet, rpn, config, two-stage
The `num_classes` should be 1 in RPN, but get {rpn_head_num_
console warning mmdet, rpn, two-stage, config
'num_dn_queries' should be set when using dynamic dn groups,
console info mmdet, dino, detr, config
Byte tensor for attn_mask is deprecated.
console info mmdet, attention, pytorch, deprecation, dtype
Default ``avg_non_ignore`` is False, if you would like to ig
console warning mmdet, loss, cross-entropy, ignore-index
DeprecationWarning: Setting "linear=True" in iou_loss is dep
console warning mmdet, deprecation, iou-loss, config
DeprecationWarning: Setting "linear=True" in IOULoss is depr
console warning mmdet, deprecation, iou-loss, constructor
Since cross entropy is not set, the num_classes will be igno
console warning mmdet, reid, config, unused-argument
``ignore_label`` and ``loss_weight`` would be deprecated soo
console warning mmdet, deprecation, semantic-head, panoptic
DeprecationWarning: pretrained is a deprecated, please use "
console warning mmdet, deprecation, init-cfg, weights
If you want to use cost-based ATSSAssigner, please set cls_
console warning mmdet, atss, assigner, ddod
``build_sampler`` would be deprecated soon, please use ``mmd
console warning mmdet, deprecation, registry, builder
``build_iou_calculator`` would be deprecated soon, please us
console warning mmdet, deprecation, registry, iou-calculator
``build_match_cost`` would be deprecated soon, please use ``
console warning mmdet, deprecation, registry, match-cost
``build_assigner`` would be deprecated soon, please use ``mm
console warning mmdet, deprecation, registry, assigner
``build_prior_generator`` would be deprecated soon, please u
console warning mmdet, deprecation, registry, prior-generator
``build_anchor_generator`` would be deprecated soon, please
console warning mmdet, deprecation, registry, anchor-generator
DeprecationWarning: onnx_delta2bbox is deprecated in the cas
console warning mmdet, deprecation, bbox-coder, onnx, batch-decode
``grid_anchors`` would be deprecated soon. Please use ``grid
console warning mmdet, deprecation, anchor-generator, priors
``single_level_grid_anchors`` would be deprecated soon. Plea
console warning mmdet, deprecation, anchor-generator, priors
DeprecationWarning: bboxes is deprecated, please use "priors
console warning mmdet, deprecation, sampling-result, priors
DeprecationWarning: pos_bboxes is deprecated, please use "po
console warning mmdet, deprecation, sampling-result, priors
DeprecationWarning: neg_bboxes is deprecated, please use "ne
console warning mmdet, deprecation, sampling-result, priors
In rpn_proposal or test_cfg, nms_thr has been moved to a dic
console warning mmdet, deprecation, nms-config, test-cfg, aug-test
X2 < X1 value in box. Swap them.
console warning mmdetection, wbf, bounding-box, coordinate-order, ensemble
Y2 < Y1 value in box. Swap them.
console warning mmdetection, wbf, bounding-box, coordinate-order, y-axis-flip
Zero area box skipped: {}.
console warning mmdetection, wbf, degenerate-box, bounding-box, data-cleaning
config is now expected to have a `runner` section, please se
console warning mmdetection, config, deprecation, migration, runner
"imgs_per_gpu" is deprecated in MMDet V2.0. Please use "samp
console warning mmdetection, config, deprecation, batch-size, migration
Got "imgs_per_gpu"={cfg.data.imgs_per_gpu} and "samples_per_
console warning mmdetection, config, conflicting-keys, deprecation, batch-size
Automatically set "samples_per_gpu"="imgs_per_gpu"={cfg.data
console info mmdetection, config, deprecation, migration, batch-size
group` is deprecated. Currently only supports NCCL backend.
console warning mmdetection, distributed, nccl, deprecation, all-reduce
Note: the "to_float" is True, you need to ensure that the be
console info mmdetection, distributed, dtype, all-reduce, numerical-precision